A lot of changes all over. Most notably probably the fbcon updates, it's
really all over the map - mostly a lot of very small fixes.
Linus
Summary of changes from v2.5.65 to v2.5.66
============================================
Adrian Bunk <[email protected]>:
o [NF/IPV6]: Remove all ipv6_ext_hdrs from ip6tables
Alan Cox <[email protected]>:
o Remove NO_VERSION from S390x exec32
o __NO_VERSION_ for ati_pcigart
o __NO_VERSION__ for used bits of dri
o __NO_VERSION__ for ftape
o Move ipmi to new struct stuff
o fix bogus C in ite_gpio
o merge lp driver for PC98xx systems
o remove ifs from ancient backcompat in mwave driver
o rio __NO_VERSION__
o newer boards put other hw at rtc + 0x08
o real time clock support for PC9800 systems
o unbreak the acquirewdt
o fc4 doesnt need __NO_VERSION__ any more
o fix all the other watchdogs Dave's changes broke the same
o fix ide-geometry bogus printk level
o remove legacy probe code
o add hd98 driver (equivalent to hd.c for old PC9800)
o clean up ht6560 legacy ide driver
o module for legacy PC9800 ide
o remove old style probe from other legacy ide
o Update ide/legacy makefile to match changes
o fix proc handling in serverworks and sc1200 ide
o fix proc handling in sis, siimageand slc90e66
o fix /proc handling in via82cxxx
o move mac-hid to C99
o remove __NO_VERSION__ from radio drivers
o remove __NO_VERSION__ from saa7134 driver
o fix GTUNER on w9966
o Fix i2o_scsi hang
o fix 3c501 typo
o remove unused ali-ircc variable
o sk98 typo fix
o typo fix for tulip
o fix pcmcia crash with hostap
o fix pcmcia __NO_VERSION__
o pnpbios doesnt want __NO_VERSION__
o fix bogus if in advansys driver
o fix time types in aha152x
o fix buffer overrun in aha1542
o fix leak in cpqfc
o gdth update from Intel
o junkfilter sym53c41
o PC9800 has a slight funny on 8250_pnp
o serial driver for PC9800 systems
o xjack memory leak fixes
o __NO_VERSION__ for autofs
o typo fix for befs
o remove __NO_VERSION__ in cifs
o typo fix for expfs
o fix fat handling of some weirder variants
o remove __NO_VERSION__ from intermezzo
o remove __NO_VERSION__ from intermezzo #2
o remove __NO_VERSION__ from jffs
o remove __NO_VERSION__ from lockd
o Add NEC PC9800 partition tables
o remove __NO_VERSION__ from procfs
o Alpha folks said my change was wrong, revert it and note the funny
o add a new dmi flag for broken pnpbios
o add another clock tick rate variant
o add headers for upd4990a rtc/clock driver
o S/390 typo fixes
o Remove i2o pci abstractions
o update i2o build rules for change
o remove __NO_VERSION__ in mtd
o kill off IDE_DEBUG, add pc9800 ide type
o update compaq idents, correct and update intel idents
o add pc9800 port types
o no arch specific headers for upd4990a
o update Achim's address
o fix typo in oom_kill
o tidy up make rpm
o fix typo in net/core/neighbour
o unless this is a backward spanish inquisition joke
o pc9800 CS4232 driver
o fix up opti92x-ad1848
o clean up es968, fix build
o fix __NO_VERSION__ in audio_syms
o fix ";" in cs46xx
o fix i810 ifs
o fix incorrect bracketing in maestro
o __NO_VERSION__ for midi_syms
o mpu401 uses __init vars during __exit
o more __NO_VERSION__ in audio
o update emu10k1 driver (SB Live, Audigy etc)
o update emu10k1 config help
o boot code for PC9800 systems
o handle exploding pnpbios
o add pc9800 setup and topology code
o Make pci-bios function ids per machine type
o arch pre/post setup for pc9800
o PC9800 system common area definition
o sysfs typo fix
o remove odd blank line and add noacpi
o ide-default driver
o __NO_VERSION__ for ide-lib
o ide-probe update
o FOr efficient non posted I/O people need to know the target
o add __ide_set_handler to fix abort race
o use new outbsync when sending commands
o rework the reset code to fix posting and races
o remove special cases from ide_proc
o update ide-tape to match changes
o printk, version etc for ide-taskfile
o ide should check dma_on
o update ide core
o update ide-cd to new changes, add abort() handlers
o update ide-disk to changes, remove all the driver ifs
o update ide-dma support
o fix tuning of alim15x3
o fix /proc for amd ide
o fix cmd640 ide locking
o fix more proc and other oddments
o add ICH5 and Centrino to PIIX4
o add ide-default to the build
o fix aec proc handling
o remove __NO_VERSION__ from bttv
o remove lots of now dead code (no features though!)
o cpia -maintainers update
o remove __NO_VERSION__ from drm
o update ide headers to match changes
o redo the n_tty fix
o make opl3sa2 build again
o clean up the mess someone merged into 3wxxx scsi
o abstract out mach_reboot for x86 platforms
o use right object for i2o_config - kernel not user copy
o add checks to pc9800 ide reserve
o ide typo fixes #3
o Merge PC9800 keyboard driver
o ide typo fixes #2
o i2o_pci is dead
o parallel port
o merge PC9800 keyboard controller chip support
o merge PC9800 mouse driver
o ide typo fixes
o eisa reports "0 device" not "0 devices"
o Fix IDE disable_irq() deadlock
o 3ware fixups from Adam Radford
o Osamu's updates to boot98
o remove NUL v NULL confusion
o Achim is now Intel (they bought GDT)
o PC98 has its own floppy (not yet merged)
o Christoph felt the IDE_HD v PC9800 was ugly
o isapnp change
o missing config bits for PC9800 audio
o Make ide use proper removal-safe list handling (removes endless
looping / hang)
Alan Stern <[email protected]>:
o USB: Update for usb-skeleton
Alexey Kuznetsov <[email protected]>:
o [NET]: hard_header reservation
o [NET]: miscellaneous fixes
o [IPSEC]: fragmentation & tcp mss calculation
o [IPV6]: Correct CHECKSUM_HW handling in tcp_v6_send_check
Allen Curtis <[email protected]>:
o PPC32: Updates for the 8260 embedded processor and the EST and TQM
boards
o PPC32: Further 8260 update; one file was missed in the previous
commit
Andrew Morton <[email protected]>:
o [NET]: Use fancy wakeups where applicable
o Fix noirqbalance
o Pass the load address into ELF_PLAT_INIT()
o remove unused block congestion code
o timer code cleanup
o timer re-addition lockup fix
o use set_current_state in fs
o use set_current_state in mm
o Fix memory leak in copy_thread
o file_list_lock contention fixes
o file->f_list locking in tty_io.c
o file_list cleanup
o file_table: remove the private freelist
o file_list: less locking
o stack reduction in drivers/char/vt_ioctl.c
o a few missing stubs for !CONFIG_MMU
o slab changes for !CONFIG_MMU
o memleak in fs/nfs/inode.c::nfs_get_sb()
o Memleak in fs/ufs/util.c
o posix timers update
o OOPS instance counters
o io-apic.c: DO_ACTION cleanup
o fix oprofile timer race
o pgd_index/pmd_index/pte_index commentary
o /proc/sysrq-trigger: trigger sysrq functions via
o Tighten CONFIG_NUMA preconditions
o Fix nfsd_symlink() failure path
o Add error checking get_disk()
o sys_nanosleep() fix
o NMI watchdog fix
o fix nanosleep() granularity bumps
o add write_seqlock to cpufreq change notifier for TSC
o cs46xx minor fixes
o Add missing put_user checks in n_tty
o Fail setup_irq for unconfigured IRQs
o raw driver: rewrite i_mapping only on final close
o raw driver: cleanups and small fixes
o slab: tune batchcounts for large objects
o Fix floppy oops on forced unload
o Make nonlinear mappings fully pageable
o filemap_populate speedup
o x86_64: support for file offsets in pte's
o ppc64 support for file file-offset-in-pte
o inode a/c/mtime modification speedup
o Implement a/c/time speedup in ext2 & ext3
o remove lock_kernel() from inode_setattr's
o speed up vm_enough_memory()
o remove lock_kernel() from readdir implementations
o __bdevname atomicity fix
o register_blkdev() fixes
o make the bdevname() API sane
o mwave oops fixes
o dev_t [1/3]: kill cdev
o dev_t [2/3] - remove MAX_CHRDEV
o dev_t [3/3]: major.h cleanups
o timer simplification
o simplify the timer lockup avoidance code
o pagecache accounting speedup
o ext3: fix use-after-free bug
o make list.h barriers smp-only
o sync_filesystems commentary and latency fix
o fix .text.exit error in OSS awe_wave.c
o Make arch-independent syscalls return long
o More syscalls-returning-long
o remove the "half of memory" limit on mlock() and
o ptrace_notify() locking
o reenable interrupts in parport code
o warning fixes
o asm-generic/tlb.h needs swap.h
o d_lookup forgotten spin_unlock()
Art Haas <[email protected]>:
o Add C99 initializers to net/ipv4/netfilter
o Add C99 initializers for net/ipv6/netfilter code
Bart De Schuymer <[email protected]>:
o [ebtables] bugfix in ebt_ip.c
o [EBTABLES] hold usage count on table module when it contains rules
Ben Collins <[email protected]>:
o [SPARC64]: Add image target
Charles Fumuso <[email protected]>:
o [XFS] fix getdents so that xfs_da_read_buf doesn't return an
EFSCORRUPTED except when there is a real problem.
Chas Williams <[email protected]>:
o [ATM]: Fix idt77252/sch_atm/pppoatm compilation
o [ATM]: cleanup nicstat, suni and idt77105
o [ATM] nicstar doesnt count all dropped pdus and powerpc fixup
o [ATM] s/uni driver overwrites 8-/16-bit mode
Christoph Hellwig <[email protected]>:
o fix waitqueue leak in devfs_d_revalidate_wait
o rempove CONFIG_DVB_DEVFS_ONLY
o make devfs_put() static to fs/devfs/base.c
o remove DEVFS_FL_REMOVABLE
o get rid of __MOD_INC_USE_COUNT/__MOD_DEC_USE_COUNT
o devfs_mk_symlink simplification
o cleanup input_register_minor
o don't include swap.h in mm.h
o convert remaining register_pcmcia_driver users
o bring back Al's devfs changes in dv1394
o devfs_mk_dir simplification
o misc devfs_register cleanups
o make devfs_alloc_unique_number private to devfs
o fix usb_devfs_handle abuse
o [XFS] remove an unused variable
o [XFS] remove VFS_DOUNMOUNT
o [XFS] time_after takes an unsigned long
o [XFS] Minor header shuffling, removing a bunch of already-included
files and allowing 2.4/2.5 to be slightly more in sync.
Dave Jones <[email protected]>:
o [CPUFREQ] Remove duplicate test. (Already done in longhaul_init())
o [CPUFREQ] Remove duplicate cpuid check from AMD powernow-k6 driver
o [CPUFREQ] Remove duplicate cpuid check from longrun driver
o [CPUFREQ] remove duplicate cpuid check from p4 clockmod driver
o [CPUFREQ] No need to export cpufreq_governor_list, so it can be
static
o fix obvious thinko
o cciss unregister cleanup
o 3ware vendor update
o Remove old DRM4.0 code
o piix compile fix for CONFIG_PROC_FS=n
o documentation for userspace access
o fix acpi write throttle seq file breakage
o Sysfs not handling show errors
o make nbd working in 2.5.x
o Several logic bugs
o Add missing intel cache descriptor
o Make cpuid driver preempt safe
o CCISS ID updates
o cyclades region handling updates from 2.4
o add another transparent bridge
o make x86 MSR driver preempt safe
o Update K6 bug URL
o plug DRM memory leak on exit paths
o x86-64: Add missing tlb flush after change_page_attr
o fix wrong return type on parisc eisa_eeprom_llseek
o fix asm constraints in ffs
o various PCI ID updates
o Add missing time initialisation to get_cramfs_inode
o bring sparc riowatchdog in sync with 2.4
o mark context switch wrmsr path unlikely
o fix decnet compile error on newer gcc's
o add support for 8 port lava octo cards to 8250_pci
o __ipv6_regen_rndid typo fix
o Bose sound support for cs4232 OSS driver
o Add __copy_from_user checks to emu10k1
o bring OSS mad16 in sync with 2.4
o back out broken bits of previous cyclades patch
o guard mad16 debug macro
David Brownell <[email protected]>:
o USB: ehci-hcd, prink tweaks
David S. Miller <[email protected]>:
o [SPARC64]: Move sparc64_init_timers into time.c
o [SPARC64]: Kill SPARC64_USE_STICK and use real timer drivers
o [SPARC64]: Fix timer quotient calcs
o [SPARC64]: Convert Cheetah scheduler tuning to use tick_ops
o [SPARC64]: Do not allow cache_decay_ticks to be zero
o [SPARC64]: Initial cut at Hummingbird STICk support
o [SPARC64]: kernel/time.c needs asm/smp.h
o [SPARC64]: Fix typo in __hbird_read_stick asm
o [SPARC64]: Fix hbtick softint_mask
o [SPARC64]: Fix __hbird_read_stick signedness, also hbird
softint_mask
o [SPARC64]: In __hbird_write_compare, write high then low part
o [SPARC64]: Make TICK comparisons wrap-around safe by using jiffies
macros
o [SPARC64]: Ignore bit 63 of Hummingbird STICK when computing
COMPARE register values
o [SPARC64]: Do a dummy write to STICK in hbird_init_tick
o [FB ATY]: CONFIG_FB_ATY needs cfbcopyarea.o
o [SPARC64]: Sanitize all TICK privileged bit handling in tick
drivers
o [SPARC64]: Clear tick_cmpr ints properly in bootup assembly
o [IPV6]: ndisc_recv_ns returns void
o [IPV6]: Undo __constant_{n,h}to{n,h}l from anycast patch
o [NET]: Kill NETIF_F_DYNALLOC, based upon ideas from Adam J. Richter
o [SPARC]: Add die_counter changes
o [SPARC64]: Fix thread_info offsets to match restart_block layout
changes
o [CRYPTO]: Include linux/errno.h as appropriate
o [IPSEC]: Export xfrm_user_policy
o [IPSEC]: net/xfrm.h needs net/sock.h
o [IPSEC]: Fix socket leak in TCP/v6 when policy lookup fails
o [SOUND]: Fix build of SBUS code in memalloc.c
o [FB SBUS]: sbuslib.c needs linux/mm.h
o [VT]: vc_pos needs to be unsigned long
o [SOUND]: Fix ioctl32 build by using compat_timespec
o [SOUND]: Fix rawmidi32 build by using compat_timespec
o [FB]: 64-bit cfbimgblt.c changes do not even build
o [FB FFB/CG6]: Fix image->data const typing
o [SOUND]: Fix timer32.c build by using compat_timespec
o [SPARC64]: Update defconfig
o [IPSEC]: Fix minor mispatch of xfrm splitup patch
o [SPARC64]: Fix typos in uptr changes
o [SCTP]: Match ipproto->handler changes for ipv6
o [COMPAT]: Fix net/compat.c build
o [SPARC64]: Implement file-offset-in-pte
David Stevens <[email protected]>:
o [IPV6]: Add anycast support
Davide Libenzi <[email protected]>:
o epoll with selectable ET/LT behaviour
Dean Roehrich <[email protected]>:
o [XFS] fix dmapi POSTCREATE event in xfs_create/xfs_mkdir
o [XFS] linvfs_file_mmap was updating the linux inode's atime twice
Dominik Brodowski <[email protected]>:
o pcmcia: check return values of driver_register
o pcmcia: add bus_type pcmcia_bus_type
o pcmcia: register drivers with bus
o pcmcia: remove single linked list of drivers
o pcmcia: convert pccard_cs driver to new registration interface
Duncan Sands <[email protected]>:
o USB speedtouch: cosmetic comment changes
o USB speedtouch: get rid of atmsar
Eric Sandeen <[email protected]>:
o [XFS] Fix indices into xfs_min/xfs_max for sysctls in 2.5.x
o [XFS] Last of the dir2 backmerges from Irix
o [XFS] Remove #if(n)def __KERNEL__ from xfs_error.c, not needed
o [XFS] Turn on random() if INDUCE_IO_ERROR is defined
o [XFS] use get/put_unaligned() to avoid unaligned accesses in the
extents code on 64-bit machines
o [XFS] Bump the reporting threshold on calls to XFS_ERROR_REPORT
which are most likely due to a simple user error.
Geert Uytterhoeven <[email protected]>:
o M68k exported symbols
o M68k ISA memory for Amiga PCMCIA
o M68k POSIX timers
o M68k: Add new kmap types
o Amiga PCMCIA Ethernet clean up
o M68k ifpsp060 updates
o M68k syscall updates
o M68k: Signal updates
o M68k heartbeat update
o M68k PAGE_SIZE warnings
o Q40: local_irq*() update
o ADB: Fix spelling of sigprocmask
o M68k Apollo I/O updates
o M68k gcc-3.2 warnings
o M68k struct page fix
o IDE_ARCH_ACK_INTR duplicate
o WD33c93 missing export
o M68k net warnings
o M68k SCSI warnings
o M68k NCR5380 SCSI updates
o Amiga serial updates
o Genrtc updates
o M68k SCSI driver updates
o Sun-3 linkfile fix
o Sun-3 memory zones
o Sun-3 first page
o Sun-3 NCR5380 SCSI warning
o Amiga NCR53c7xx SCSI: use z_ioremap()
o Amifb wrong interrupt
o Amiga RTC updates
o wd33c93 SCSI merge error
o Sun-3 NCR5380 SCSI printk tags
o M68k core spelling fixes
o Affs sizeof()
o M68k timekeeping update
o console_initcall() return type
o Port amifb to new fbdev API
o Amiflop mod_timer()
Geoffrey Wehrman <[email protected]>:
o [XFS] Fix target_linkzero calculation
Glen Overby <[email protected]>:
o [XFS] Add error reporting calls in error paths that return
EFSCORRUPTED
o [XFS] Fix freespace entry search to handle holes in the freespace
area correctly
o [XFS] Rearrange leaf space allocation
o [XFS] getdents fix for dir v2
o [XFS] Add stack trace print to xfs_error_report, warning cleanup
o [XFS] In open, check the inode for extents after getting the shared
lock on the inode. The inode could have changed since before the
lock.
Greg Kroah-Hartman <[email protected]>:
o i2c i2c-i801.c: remove #ifdefs and fix all printk() to use dev_*()
o i2c i2c-i801.c: remove check_region() usage
o i2c i2c-i801.c: fix up the pci id matching, and change to use
proper pci ids
o i2c i2c-i801.c: fix up formatting and whitespace issues
o i2c i2c-piix4.c: remove check_region() call
o i2c i2c-piix4: remove #ifdefs and fix all printk() to use dev_*()
o i2c i2c-piix4.c: fix up formatting and whitespace issues
o i2c i2c-ali15x3.c: remove #ifdefs and fix all printk() to use
dev_*()
o i2c i2c-ali15x3.c: remove check_region() call
o i2c i2c-ali15x3.c: fix up formatting and whitespace issues
o i2c i2c-amd756.c: remove some #ifdefs and fix all printk() to use
dev_*()
o i2c i2c-amd8111.c: change a few printk() to dev_warn()
o i2c i2c-amd8111.c: change the pci driver name to have "2" in it
based on previous comments
o i2c: added i2c-isa bus controller driver
o i2c: add initial driver model support for i2c drivers
o USB: whiteheat bugfix (bugzilla.kernel.org #314)
o USB: pegasus: fix up GFP_DMA usages. (bugzilla.kernel.org #418)
Henning Meier-Geinitz <[email protected]>:
o USB: new ids for scanner driver
Hideaki Yoshifuji <[email protected]>:
o [IPV6]: Use RFC2553 constant variable
o [IPSEC]: Fix obvious typo in xfrm_sk_clone_policy
o [IPSEC]: Split up XFRM Subsystem
o [IPV6]: Use "const" qualifier
o [IPV6]: Use ipv6_addr_any() for testing unspecified address
o [IPV6]: Fix BUG 483, do not call crypto_alloc_tfm from illegal
context
o [IPSEC]: Fix bug in xfrm_parse_spi()
Ingo Molnar <[email protected]>:
o mm/swapfile.c manual reschedule
o posix-timers-cleanup-2.5.65-A5
Ivan Kokshaysky <[email protected]>:
o [PCI] Make setup-bus.c aware of cardbus bridges
o [PCI] Fix incorrect PCI cache line size assumptions
o [PCI] Don't call pci_update_resource() for bridge resources
James Morris <[email protected]>:
o [IPSEC]: fix skb leak in ah and esp
o [IPSEC]: return error when no dst in ah & esp output
o [IPSEC]: Fix parsing of 16-bit ipcomp cpi
o [IPSEC] Add initial compression support for pfkey and xfrm_algo
o [CRYPTO]: Make use of crypto_exit_ops() during crypto_free_tfm()
James Simmons <jsimmons@kozmo.(none)>:
o [NEOMAGIC FBDEV] Fix to work with no 21xx versions of the chip
o [FBCON] More struct display cleanup. Also killed off static buffer
for accel_putcs
o [ATY128 FBDEV] Moved aty128fb to aty/ and a few minor changes so
aty128fb.c compiles with the newest compiler standards
o [FBDEV] Enhanced data buffer management for drawing very large
images
o [FBDEV] Menu cleanups. Added in depenedency needed. More cleanup in
fbcon layer
o [CONSOLE] Nuked a few gloabl variables. Now that the console system
supports different sized screens these gloabl variables are a bad
idea
o [FBCON] Killing off more cursor fields in struct display. Use what
is in struct vc_data
o [FBCON] Cursor handling clean up. I nuked several static variables
o [FBCON]More optimizations. Removed moving struct display around
o [FBDEV] Killed of a static buffer in the generic software cursor.
We didn't need it and it is a bad idea to have a static buffer is
we have more than one framebuffer
o [AMIGA FBDEV] Ported over to new api
o [CONTROL/PLATNIUM FBDEV] Small cleanups to latest changes
o [FBCON] Nuked the final gloabl variables for the cursor code
o [GENERIC ACCELERATION] Fixed the generic image drawing function
tfor 64 bit machines
o Accel rountines pass in constant data into each function. The
reason being was some of the code in the upper layers depended on
the data being passed to the low level function not be altered
because the upper layers was altering the data themselves
o [RADEON FBDEV] Add cursor support. Now the cursor is back
o [RIVA FBDEV] SUpprot Directcolor mode. Needed for some cards
o [ATY FBDEV] Updates to support Rage Mobility Chipstes
o [ATY FBDEV] Reversed mobilty patches. They busted every other card
o [FBCON] Removal of useless code
o [LOGO] New better logo code
o [FBDEV] Data in struct fb_image is now const
o Removed obsolete functions in fbcon.c and re-enabled mapping
console(s) to a framebuffer device. A few compile fixes for rivafb
and using standard macros for vgacon.c
o [FBDEEV] Need to add support to build pnmtologo
o [FBDEV] Minor fixes for logo code
o [ATY FBDEV] Rage XL cards can now be booted with needed the BIOS
:-)
o [FRAMEBUFFER]: cfbcopyarea accesses fb without using
FB_{READ,WRITE}L
o [GENRIC ACCEL] Megred David Millers changes with my own
o [FBDEV] Accelerated functions pass in const structs
o [FBDEV] Updates for the SIS fbdev driver to the new api. Removed
poll. We wil use signals in the future instead
o [FBCON] Help clear margins for modes where the resolution does
quite fit the console size
o [TWIN TURBO FBDEV] Ported over to new api
o [SIS FBDEV] Make it compile as a module
o [SIS FBDEV] More sisfb driver updates
o [FBDEV] Standardized using xxfb for setup strings
o [SIS FBDEV] Added Maintiner for SIS fbdev driver
o [FBDEV] If a colormap contains no transparency information,
fb_set_cmap() calls fb_setcolreg() with trans = 0. This causes all
CLUT entries to be fully transparent on hardware that does have
transparency information in the CLUT registers.
o [FBDEV] Ug!!! For some reason BK keeps removing this change. I hope
this is the last time I have to add it
Jaroslav Kysela <[email protected]>:
o ALSA update
o ALSA update (0.9.2)
Jason McMullan <[email protected]>:
o USB HID: Ignore P5 Data Glove (2.4 and 2.5 patches)
Jeb Cramer <[email protected]>:
o [E1000] Documentation/networking/e1000.txt updates
o [E1000] Version, copyright, changelog and MAINTAINERS
o [E1000] Spd/dplx abstraction; eeprom size changes
o [E1000] IRQ registration fix
o [E1000] Added 82541 & 82547 support
o [E1000] Added MII support
o [E1000] Modulus math removed
o [E1000] Perform single PCI read per interrupt
o [E1000] Tx Descriptor cleanup
o [E1000] Read/Write register macro optimizations
o [E1000] Compaq to HP branding change
o [E1000] Whitespace changes
o [E1000] Added Tx FIFO flush routine
o [E1000] Added Interrupt Throttle Rate tuning support
o [E1000] Controller wake-up thru ASF fix
o [E1000] whitespace fix from previous patches
o [E1000] NAPI re-insertion w/ changes
Jeff Garzik <[email protected]>:
o [netdrvr tg3] fix memleak in DMA test
o [via-rhine] note that Roger is maintainer, in MAINTAINERS
Jens Axboe <[email protected]>:
o cdrom buffer too small
o Check remailing length in ide-cd.c correctly
Jon Grimm <[email protected]>:
o [IPSEC]: Fix SKB alloc len in ip6_build_xmit
o [SCTP] Add SCTP_SET_PEER_PRIMARY get/setsockopt
o [SCTP] Supported address types should be based on pf_family
o [SCTP] Only consider C-E bundling up until C-E has been sent
o [SCTP] Fix typo in tsnmap.c (Norbert Kiesel)
o [SCTP] Add SCTP_NODELAY sockopt and message delay (ardelle.fan)
o [SCTP] Fix out_qlen (pending data) count
o [SCTP] Receiver SWS prevention
o [SCTP] Fix panic on close()
Jonathan Corbet <[email protected]>:
o Request queue micropatch
Joshua Uziel <[email protected]>:
o [SPARC64]: Fix up_clock_tick export
KANDA Mitsuru <[email protected]>:
o [IPV6]: Process all extension headers via ipproto->handler
Linus Torvalds <[email protected]>:
o Fix a rather theoretical race if an NMI happens when a debug fault
happens exactly on the sysenter entry path before the kernel stacks
have been switched to the proper ones.
o Fix sound driver timeout types. Again
o Avoid warning with modern gcc's in xfrm_policy.c
o Alan broke the build. Fix it thusly
o Use '#ifdef' instead of '#if' to test CONFIG_xxxx variables. It
works both ways, but anal compilers will warn about using
identifiers that have never been defined in preprocessor expression
statements.
o Use '#ifdef' to check for CONFIG_xxx definitions
o Avoid using the gcc-ism of creating an anonymous structure directly
by having a cast followed by an initializer. It seems even gcc
can't do it right anyway in some versions (as reported by Jens
Axboe).
Luis F. Ortiz <[email protected]>:
o [SPARC64]: Define IDE MAX_HWIFS like x86
Matthew Wilcox <[email protected]>:
o PARISC update
o [NET]: Optimize handling of CONFIG_NET=n
Nathan Scott <[email protected]>:
o [XFS] First stage of behavior code cleanup - removes a bunch of
unused macros related to read/write locking the behavior change.
o [XFS] Export end_buffer_async_write, needed for unwritten extent
support in XFS
o [XFS] Implement support for unwritten extents in XFS
o [XFS] Find more appropriate homes for uuid_t, timespec_t and
xfs_dirent_t defs
o [XFS] Remove unneeded initialisations to zero, formatting cleanups,
remove a no-longer-correct-comment, fix up symlink error path code,
several minor changes to help keep this code more in sync with 2.4.
o [XFS] Fix permission checks for some ioctls. Its now possible for
ordinary users to use the preallocation calls if unwritten extents
are enabled, and a couple of places where we were allowing
operations if unwritten extents are enabled, but shouldn't have
been, have been closed up.
Neil Brown <[email protected]>:
o Fix a few MD bugs
o Remove obsolete NFSD syscall varients
Oleg Drokin <[email protected]>:
o USB: Memleak in drivers/usb/hub.c::usb_reset_device
Oliver Neukum <[email protected]>:
o USB: fix to synchronous API regarding memory allocation
Osamu Tomita <[email protected]>:
o Support PC-9800 subarchitecture (9/14) NIC
Paul Mackerras <[email protected]>:
o PPC32: Make a ppc32 version of pcibios_resource_to_bus, which adds
an offset where needed.
o PPC32: Makefile tidy-up, mostly from Sam Ravnborg
o PPC32: Fix asm/rtc.h so drivers/char/genrtc.c compiles
o PPC32: Convert uses of ide_ioreg_t to unsigned long
o PPC32: Fix ide_init_hwif_ports for powermac
o update macintosh-specific headers
o update via-cuda driver
o SMP-safe macserial driver
o update via-pmu driver
o update MESH scsi driver
o update mac53c94 scsi driver
o fix powerbook media bay
Petr Vandrovec <[email protected]>:
o Fix ncpfs and rpcgss order in fs/Kconfig
Randy Dunlap <[email protected]>:
o USB: reduce stack usage in cdc-ether
o reduce stack in cdrom/optcd.c
o reduce stack in wireless/airo.c
Russell King <[email protected]>:
o [ARM] Minor updates/fixes to ARM PCI support code
o [ARM] Fix timeouts to use the correct type
o [ARM] Remove explicit IRQ disable/enable in PXA timer IRQ
o [ARM] Add typechecking to local_irq_save()
o [ARM] Update CLPS7500 support
o [ARM] Distinguish between the various oops messages better
o [ARM] Ensure transmitter starts before leaving ssp_write_word()
o [ARM] Fix more timeouts to use correct type
o [ARM] Add L1_CACHE_SHIFT to asm-arm/cache.h
o [ARM] Update Acorn SCSI drivers
o [ARM] Fix CONFIG_CPU_FREQ_GOV_USERSPACE warning
o [ARM] Remove head-netwinder.S
Scott Feldman <[email protected]>:
o [E100] back out memleak patch cuz it messed up following
o [E100] Update Documentation/networking/e100.txt
o [E100] update version, copyright year, changelog
o [E100] Spelling mistakes
o [E100] Add support for VLAN hw offload
o [E100] Clean up #include order
o [E100] Bug fix on setting up Tx csum
o [E100] Banish strong branding marketing strings
o [E100] forced speed/duplex link recover
o [E100] ICH5 support added
o [E100] Honor WOL settings in EEPROM
o [E100] interrupt handler free fix
o [E100] Validate updates to MAC address
o [E100] ethtool EEPROM and GSTRING fixes
o [E100] ASF wakeup enabled, but only if set in EEPROM
o [E1000] Increase default Rx descriptors to 256
Sridhar Samudrala <[email protected]>:
o [SCTP] accept() support for TCP-style SCTP sockets
o [SCTP] Minor changes for tcp-style socket support
Stephen Lord <[email protected]>:
o [XFS] initialize the iovec length in symlink cases
o [XFS] If cmn_err is called with a CR character at the end of the
input then do not add one.
o [XFS] remove _KERNEL from the flags used to turn macros into
functions
o [XFS] rework readdir to be closer to the irix model internally, do
all the filldir fixup at the linvfs layer. This is the V2 directory
component, the V1 code still needs fixing up. We now return the
same directory offsets as Irix does.
o [XFS] validate_fields is called on a vnode to update directory
related fields after a create/remove etc. Make sure we pass in all
the flags for the status fields we want. NBLOCKS was missing and
working by accident.
o [XFS] prevent readdir from returning offsets of more than 2^31,
these confuse user space. This limits the maximum amount of names
in a directory on linux to 2Gbytes, which should not be a problem.
o [XFS] remove a couple more sync transactions from xfs
o [XFS] fix a couple of kmem issues, check for OOM in kmem_relalloc
more and in the out of mem case, panic in the sleep case, not the
non-sleep case.
o [XFS] remove bad debug code
o [XFS] move vn_alloc stat from xfs_iget to vn_initialize
o [XFS] reduce byte swapping and spinlock usage in log write path
o [XFS] remove some unbounded loops from the unwritten and unmapped
page processing code. As files get larger, these code paths have
the potential to hog the cpu for long periods of time. Just cap the
unmapped page case, and the unwritten one is supposed to be
stopping at the end of the extent anyway.
Stephen Rothwell <[email protected]>:
o compat_uptr_t and compat_ptr
o [COMPAT] cleanups in net/compat.c and related files
o fix typo in compat_ptr
o another fix for compat_ptr
Steven Cole <[email protected]>:
o i2c: spelling corrections for drivers/i2c
Thomas Walpuski <[email protected]>:
o [IPSEC]: Fix null authentication/encryption
Tom Lendacky <[email protected]>:
o [IPSEC]: Add IPV6_{IPSEC,XFRM}_POLICY socket option support
o [IPSEC]: IPV6 source address not set correctly in xfrm_state
Ulrich Drepper <[email protected]>:
o more generic syscall return value type fixes
Mirrored at cvs -d:pserver:[email protected]:/home/cvs in the
linux-2.5 sub directory.
--
---
Larry McVoy lm at bitmover.com http://www.bitmover.com/lm
I encountered a minor build failure in fs/cramfs, which I was able to
fix with the following patch. I'm not sure if it's exactly the right
fix - does the original author really mean to have zeroes for those
values? Anyway, please consider applying to 2.5.66 if it's okay.
Jeremy Brown
--- linux-2.5.66/fs/cramfs/inode.c.orig 2003-03-24 16:01:07.000000000 -0800
+++ linux-2.5.66/fs/cramfs/inode.c 2003-03-24 16:09:18.000000000 -0800
@@ -43,6 +43,7 @@
static struct inode *get_cramfs_inode(struct super_block *sb, struct cramfs_inode * cramfs_inode)
{
struct inode * inode = new_inode(sb);
+ struct timespec zerotime = {0, 0};
if (inode) {
inode->i_mode = cramfs_inode->mode;
@@ -51,7 +52,7 @@
inode->i_blocks = (cramfs_inode->size - 1) / 512 + 1;
inode->i_blksize = PAGE_CACHE_SIZE;
inode->i_gid = cramfs_inode->gid;
- inode->i_mtime = inode->i_atime = inode->i_ctime = 0;
+ inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime;
inode->i_ino = CRAMINO(cramfs_inode);
/* inode->i_nlink is left 1 - arguably wrong for directories,
but it's the best we can do without reading the directory
drivers/char/genrtc.c:100: warning: static declaration for
`gen_rtc_interrupt' follows non-static
drivers/char/genrtc.c: In function `gen_rtc_timer':
drivers/char/genrtc.c:135: warning: comparison of distinct pointer
types lacks a cast
drivers/char/genrtc.c: In function `gen_rtc_open':
drivers/char/genrtc.c:358: warning: `MOD_INC_USE_COUNT' is deprecated
(declared
at include/linux/module.h:431)
drivers/char/genrtc.c: In function `gen_rtc_release':
drivers/char/genrtc.c:377: warning: `MOD_DEC_USE_COUNT' is deprecated
(declared
at include/linux/module.h:443)
drivers/char/genrtc.c: In function `gen_rtc_proc_output':
drivers/char/genrtc.c:453: void value not ignored as it ought to be
drivers/char/genrtc.c:498: `RTC_BATT_BAD' undeclared (first use in
this function)
drivers/char/genrtc.c:498: (Each undeclared identifier is reported
only once
drivers/char/genrtc.c:498: for each function it appears in.)
florin
--
"NT is to UNIX what a doughnut is to a particle accelerator."
On Mon, 24 Mar 2003 15:26:47 -0800 (PST) Linus Torvalds (LT) wrote:
LT> Summary of changes from v2.5.65 to v2.5.66
LT> ============================================
LT> Greg Kroah-Hartman <[email protected]>:
LT> o i2c i2c-i801.c: remove #ifdefs and fix all printk() to use dev_*()
LT> o i2c i2c-i801.c: remove check_region() usage
LT> o i2c i2c-i801.c: fix up the pci id matching, and change to use
LT> proper pci ids
LT> o i2c i2c-i801.c: fix up formatting and whitespace issues
LT> o i2c i2c-piix4.c: remove check_region() call
LT> o i2c i2c-piix4: remove #ifdefs and fix all printk() to use dev_*()
LT> o i2c i2c-piix4.c: fix up formatting and whitespace issues
LT> o i2c i2c-ali15x3.c: remove #ifdefs and fix all printk() to use
LT> dev_*()
LT> o i2c i2c-ali15x3.c: remove check_region() call
LT> o i2c i2c-ali15x3.c: fix up formatting and whitespace issues
LT> o i2c i2c-amd756.c: remove some #ifdefs and fix all printk() to use
LT> dev_*()
LT> o i2c i2c-amd8111.c: change a few printk() to dev_warn()
LT> o i2c i2c-amd8111.c: change the pci driver name to have "2" in it
LT> based on previous comments
LT> o i2c: added i2c-isa bus controller driver
LT> o i2c: add initial driver model support for i2c drivers
LT> o USB: whiteheat bugfix (bugzilla.kernel.org #314)
LT> o USB: pegasus: fix up GFP_DMA usages. (bugzilla.kernel.org #418)
Hi,
I guess it's one of the I2C changes which breaks 2.5.66 and bttv, because
2.5.65 was still ok and there don't seem to be any relevant bttv changes.
With 2.5.66 I get a kernel oops with the following backtrace:
kobject_init + 0x2d/0x50
kobject_register + 0x17/0x70
get_bus + 0x1d/0x40
bus_add_driver + 0x5b/0xe0
driver_register + 0x2f/0x40
i2c_add_driver + 0x85/0xf0
bttv_init_module + 0x93/0xf0
msp3400_init_module + 0xf/0x20
init + 0x33/0x190
init + 0x0/0x190
kernel_thread_helper + 0x5/0x18
EIP is at kobject_get + 0x13/0x50
Code: 8b 43 10 85 c0 ...
Oops copied by hand, so I hope it's correct.
-Udo.
On Tue, Mar 25, 2003 at 01:22:52AM +0100, Udo A. Steinberg wrote:
> On Mon, 24 Mar 2003 15:26:47 -0800 (PST) Linus Torvalds (LT) wrote:
>
> LT> Summary of changes from v2.5.65 to v2.5.66
> LT> ============================================
> LT> Greg Kroah-Hartman <[email protected]>:
> LT> o i2c i2c-i801.c: remove #ifdefs and fix all printk() to use dev_*()
> LT> o i2c i2c-i801.c: remove check_region() usage
> LT> o i2c i2c-i801.c: fix up the pci id matching, and change to use
> LT> proper pci ids
> LT> o i2c i2c-i801.c: fix up formatting and whitespace issues
> LT> o i2c i2c-piix4.c: remove check_region() call
> LT> o i2c i2c-piix4: remove #ifdefs and fix all printk() to use dev_*()
> LT> o i2c i2c-piix4.c: fix up formatting and whitespace issues
> LT> o i2c i2c-ali15x3.c: remove #ifdefs and fix all printk() to use
> LT> dev_*()
> LT> o i2c i2c-ali15x3.c: remove check_region() call
> LT> o i2c i2c-ali15x3.c: fix up formatting and whitespace issues
> LT> o i2c i2c-amd756.c: remove some #ifdefs and fix all printk() to use
> LT> dev_*()
> LT> o i2c i2c-amd8111.c: change a few printk() to dev_warn()
> LT> o i2c i2c-amd8111.c: change the pci driver name to have "2" in it
> LT> based on previous comments
> LT> o i2c: added i2c-isa bus controller driver
> LT> o i2c: add initial driver model support for i2c drivers
> LT> o USB: whiteheat bugfix (bugzilla.kernel.org #314)
> LT> o USB: pegasus: fix up GFP_DMA usages. (bugzilla.kernel.org #418)
>
> Hi,
>
> I guess it's one of the I2C changes which breaks 2.5.66 and bttv, because
> 2.5.65 was still ok and there don't seem to be any relevant bttv changes.
>
> With 2.5.66 I get a kernel oops with the following backtrace:
Yes, I sent out some patches a few evenings ago to lkml that should fix
this problem. I'm resyncing them with 2.5.66 right now and will send
them to Linus in a bit.
If you want to get around this for now, in the bttv driver, memset the
i2c_client structure to 0 after it is initialized. That will solve the
problem.
Hope this helps,
greg k-h
Compile statistics: 2.5.66
2.5.65 2.5.66
-------------------- -----------------
bzImage (defconfig) 14 warnings 8 warnings
0 errors 0 errors
bzImage (allmodconfig) 30 warnings 11 warnings
12 errors 6 errors
modules (allmodconfig) 2421 warnings 2285 warnings
100 errors 96 errors
Compile statistics have been for kernel releases from 2.5.46 to 2.5.66
at: http://www.osdl.org/archive/cherry/stability (will be updated by 6PM PST).
Other stability-related links:
OSDL Stability page:
http://osdl.org/projects/26lnxstblztn/results/
Nightly linux-2.5 bk build:
http://www.osdl.org/archive/cherry/stability/linus-tree/running.txt
2.5 porting items:
http://www.osdl.org/archive/cherry/stability/linus-tree/port_items.txt
2.5 porting items history:
http://www.osdl.org/archive/cherry/stability/linus-tree/port_history.txt
John
On Mon, 24 Mar 2003 16:30:48 -0800 Greg KH (GK) wrote:
Hello,
GK> Yes, I sent out some patches a few evenings ago to lkml that should fix
GK> this problem. I'm resyncing them with 2.5.66 right now and will send
GK> them to Linus in a bit.
I've found all 13 patches and applied them here.
GK> If you want to get around this for now, in the bttv driver, memset the
GK> i2c_client structure to 0 after it is initialized. That will solve the
GK> problem.
Yes, the oops is cured now.
Thanks.
-Udo.
On Tue, Mar 25, 2003 at 02:18:25AM +0100, Udo A. Steinberg wrote:
> On Mon, 24 Mar 2003 16:30:48 -0800 Greg KH (GK) wrote:
>
> GK> Yes, I sent out some patches a few evenings ago to lkml that should fix
> GK> this problem. I'm resyncing them with 2.5.66 right now and will send
> GK> them to Linus in a bit.
>
> I've found all 13 patches and applied them here.
>
> GK> If you want to get around this for now, in the bttv driver, memset the
> GK> i2c_client structure to 0 after it is initialized. That will solve the
> GK> problem.
>
> Yes, the oops is cured now.
Great, thanks for letting me know.
greg k-h
I've encountered two reproducable bugs, and one feature which can and can not
be called bug:
1. Radeon frame buffer driver doesn't support mode change in kernel boot
params. In 2.6.65 it is OK.
append line from lilo.conf
append = " hdd=ide-scsi video=radeon:1024x768-24@100"
No mether what is in video=radeon:..., resolution is always set to 80x30 with
60Hz refresh.
2. Radeon frame buffer mode switching gives unexpected results. When switching
from lower res to higher, switching is ok but you still have old chararcter
res. eg. 80x30. The text is located in upper left corner, and the right side
off the text area is filled with garbage. Bellow text area there is nothing.
3. Cursor disapears when moving with cursor keys. This is very annoying when
you are editing text for example.
My config is attached, gcc version is 2.95.3, modutils 2.4.21.
--
Regards,
Tanasković Toplica
> I've encountered two reproducable bugs, and one feature which can and can not
> be called bug:
>
> 1. Radeon frame buffer driver doesn't support mode change in kernel boot
> params. In 2.6.65 it is OK.
>
> append line from lilo.conf
>
> append = " hdd=ide-scsi video=radeon:1024x768-24@100"
>
> No mether what is in video=radeon:..., resolution is always set to 80x30 with
> 60Hz refresh.
Its radeonfb now instead of radeon. Alot of drivers where broken in this
way. Now every driver follows a standard.
> 2. Radeon frame buffer mode switching gives unexpected results. When switching
> from lower res to higher, switching is ok but you still have old chararcter
> res. eg. 80x30. The text is located in upper left corner, and the right side
> off the text area is filled with garbage. Bellow text area there is nothing.
For console resizing try using stty cols xxx rows xx.
Fbset hasn't been updated to the new changes yet. I plan to fix up fbset
and also maybe stty. The tty layer can work on the pixel level. See struct
winsize in asm/termios.h
> 3. Cursor disapears when moving with cursor keys. This is very annoying when
> you are editing text for example.
>
> My config is attached, gcc version is 2.95.3, modutils 2.4.21.
I didn't notice this. I will track it down.
On Mon, Mar 24, 2003 at 04:30:48PM -0800, Greg KH wrote:
> Yes, I sent out some patches a few evenings ago to lkml that should fix
> this problem. I'm resyncing them with 2.5.66 right now and will send
> them to Linus in a bit.
I have an oops of my very own to report, and this one is with the afore
mentioned patches applied:
mice: PS/2 mouse device common for all mice
logimb.c: Didn't find Logitech busmouse at 0x23c
input: PC Speaker
input: PS/2 Synaptics Touchpad on isa0060/serio1
serio: i8042 AUX port at 0x60,0x64 irq 12
input: AT Set 2 keyboard on isa0060/serio0
serio: i8042 KBD port at 0x60,0x64 irq 1
i2c-dev.o: i2c /dev entries driver module version 2.7.0 (20021208)
i2c-proc.o version 2.7.0 (20021208)
i2c-pixx4 version 2.7.0 (20021208)
piix4 smbus 00.07.3: Found Intel Corp. 82371AB/EM/MB PIIX4 device
Unable to handle kernel NULL pointer dereference at vertual address 00000000
printing eip:
c02fb031
*pde = 00000000
Oops: 0000 [#1]
CPU: 0
EIP: 0060:[<c02fb831>] Not tainted
EFLAGS: 00010202
EIP is at piix4_transaction+0x101/0x15c
eax: 00000000 ebx: 00000004 ecx: 00000000 edx: ffffffff
esi: 00000000 edi: ffffffff ebp: c1299ee8 esp: c1299ed8
ds: 007b es: 007b ss: 0068
Process swapper (pid: 1, threadinfo=c1298000 task=c12fe040)
Stack: c04cfdb6 00000000 00000000 00000000 c1299f00 c02fba7e 00000018 00000000
00000000 00000648 c1299f4c c02f9a3b c04cfc60 00000018 00000000 00000000
00000000 00000000 00000000 00000000 00000001 00000001 c04cfc78 0000ffff
Call Trace:
[<c02fba7e>] piix4_access+0x1f2/0x2d0
[<c02f9a3b>] i2c_smbus_xfer+0x123/0x1b8
[<c02fb43c>] i2c_detect+0x44/0x494
[<c02fbbe6>] adm1021_attach_adapter+0x16/0x1c
[<c02fbbec>] adm1021_detect+0x0/0x2d8
[<c02f7f31>] i2x_add_driver+0xdd/0x104
[<c010502c>] init+0x0/0x144
[<c0105049>] init+0x1d/0x144
[<c010502c>] init+0x0/0x144
[<c0107211>] kernel_thread_helper+0x5/0xc
Code: 8b 00 50 68 20 36 44 c0 e8 fa ff e1 ff 83 c4 0c 0f b7 15 08
<0>Kernel panic: Attempted to kill init!
After a long pause it then looped, once per second with the call trace
repeating over and over again in the framebufer. Basically every time the
cursor was blinked. The end of these call traces is also the above call
trace. Also, the call trace was looping due to the code being called in
an illegal context in mm/slab.c:1723.
--
"Other countries of course, bear the same risk. But there's no doubt his
hatred is mainly directed at us. After all this is the guy who tried to
kill my dad."
- George W. Bush Jr, Leader of the United States Regime
September 26, 2002 (from a political fundraiser in Houston, Texas)
On Tue, Mar 25, 2003 at 03:18:03PM +1100, CaT wrote:
> On Mon, Mar 24, 2003 at 04:30:48PM -0800, Greg KH wrote:
> > Yes, I sent out some patches a few evenings ago to lkml that should fix
> > this problem. I'm resyncing them with 2.5.66 right now and will send
> > them to Linus in a bit.
>
> I have an oops of my very own to report, and this one is with the afore
> mentioned patches applied:
>
> mice: PS/2 mouse device common for all mice
> logimb.c: Didn't find Logitech busmouse at 0x23c
> input: PC Speaker
> input: PS/2 Synaptics Touchpad on isa0060/serio1
> serio: i8042 AUX port at 0x60,0x64 irq 12
> input: AT Set 2 keyboard on isa0060/serio0
> serio: i8042 KBD port at 0x60,0x64 irq 1
> i2c-dev.o: i2c /dev entries driver module version 2.7.0 (20021208)
> i2c-proc.o version 2.7.0 (20021208)
> i2c-pixx4 version 2.7.0 (20021208)
> piix4 smbus 00.07.3: Found Intel Corp. 82371AB/EM/MB PIIX4 device
> Unable to handle kernel NULL pointer dereference at vertual address 00000000
Do you have the patches I just sent out a few hours ago?
You will need the last one, I've attached it here. Let me know if it
fixes this or not.
thanks,
greg k-h
ChangeSet 1.985.1.4, 2003/03/24 15:16:12-08:00, [email protected]
[PATCH] i2c: set up a "generic" i2c driver to prevent oopses when devices are registering.
This is needed as we are still not using the driver core model for
matching up devices to drivers, but doing it by hand. Once that is
changed, this will not be needed.
drivers/i2c/i2c-core.c | 9 +++++++++
1 files changed, 9 insertions(+)
diff -Nru a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c
--- a/drivers/i2c/i2c-core.c Mon Mar 24 17:26:51 2003
+++ b/drivers/i2c/i2c-core.c Mon Mar 24 17:26:51 2003
@@ -65,6 +65,14 @@
return 0;
}
+static struct device_driver i2c_generic_driver = {
+ .name = "i2c",
+ .bus = &i2c_bus_type,
+ .probe = i2c_device_probe,
+ .remove = i2c_device_remove,
+};
+
+
/* ---------------------------------------------------
* registering functions
* ---------------------------------------------------
@@ -106,6 +114,7 @@
if (adap->dev.parent == NULL)
adap->dev.parent = &legacy_bus;
sprintf(adap->dev.bus_id, "i2c-%d", i);
+ adap->dev.driver = &i2c_generic_driver;
device_register(&adap->dev);
/* inform drivers of new adapters */
Good! But I don't see my compilation fixes. Will resend.
Below a new one.
Andries
diff -u --recursive --new-file -X /linux/dontdiff a/fs/partitions/msdos.c b/fs/partitions/msdos.c
--- a/fs/partitions/msdos.c Mon Feb 24 23:02:56 2003
+++ b/fs/partitions/msdos.c Tue Mar 25 06:22:31 2003
@@ -219,7 +219,7 @@
* Create devices for BSD partitions listed in a disklabel, under a
* dos-like partition. See parse_extended() for more information.
*/
-static void
+void
parse_bsd(struct parsed_partitions *state, struct block_device *bdev,
u32 offset, u32 size, int origin, char *flavour,
int max_partitions)
On Mon, Mar 24, 2003 at 08:34:54PM -0800, Greg KH wrote:
> Do you have the patches I just sent out a few hours ago?
I hadn't applied those (only the ones for 2.5.65-bk?) but I have now
and currently my cpu temp is a pleasant 37 degrees (I think... I have
no idea what temp1 and temp2 really mean :)
> You will need the last one, I've attached it here. Let me know if it
> fixes this or not.
I grabbed the entire patchset you posted for 2.5.66 and applied. Booted
without oops. Me happy. :)
Thanks.
--
"Other countries of course, bear the same risk. But there's no doubt his
hatred is mainly directed at us. After all this is the guy who tried to
kill my dad."
- George W. Bush Jr, Leader of the United States Regime
September 26, 2002 (from a political fundraiser in Houston, Texas)
On Tue, Mar 25, 2003 at 04:56:13PM +1100, CaT wrote:
> > You will need the last one, I've attached it here. Let me know if it
> > fixes this or not.
>
> I grabbed the entire patchset you posted for 2.5.66 and applied. Booted
> without oops. Me happy. :)
Nice, thanks for testing, I really appreciate it.
greg k-h
On Mon, Mar 24, 2003 at 10:51:25PM -0800, Greg KH wrote:
> On Tue, Mar 25, 2003 at 04:56:13PM +1100, CaT wrote:
> > > You will need the last one, I've attached it here. Let me know if it
> > > fixes this or not.
> >
> > I grabbed the entire patchset you posted for 2.5.66 and applied. Booted
> > without oops. Me happy. :)
>
> Nice, thanks for testing, I really appreciate it.
No worries. Glad to help. :)
One point of note though. Should this be happening?
i2c-dev.o: i2c /dev entries driver module version 2.7.0 (20021208)
i2c-proc.o version 2.7.0 (20021208)
i2c-piix4 version 2.7.0 (20021208)
piix4 smbus 00:07.3: Found Intel Corp. 82371AB/EB/MB PIIX4 device
i2c i2c-0: Error: no response!
i2c i2c-0: Error: no response!
i2c i2c-0: Error: no response!
i2c i2c-0: Error: no response!
i2c i2c-0: Error: no response!
i2c i2c-0: Error: no response!
i2c i2c-0: Error: no response!
registering i2c_dev_0
i2c i2c-0: Error: no response!
I only have the following compiled in:
# I2C support
CONFIG_I2C=y
CONFIG_I2C_CHARDEV=y
CONFIG_I2C_PROC=y
CONFIG_I2C_PIIX4=y
CONFIG_SENSORS_ADM1021=y
I trimmed things down to this through testing (I compiled in many
options and then left only the ones that appeared in
/proc/sys/dev/sensors)
--
"Other countries of course, bear the same risk. But there's no doubt his
hatred is mainly directed at us. After all this is the guy who tried to
kill my dad."
- George W. Bush Jr, Leader of the United States Regime
September 26, 2002 (from a political fundraiser in Houston, Texas)
On Tue, Mar 25, 2003 at 06:22:30PM +1100, CaT wrote:
> On Mon, Mar 24, 2003 at 10:51:25PM -0800, Greg KH wrote:
> > On Tue, Mar 25, 2003 at 04:56:13PM +1100, CaT wrote:
> > > > You will need the last one, I've attached it here. Let me know if it
> > > > fixes this or not.
> > >
> > > I grabbed the entire patchset you posted for 2.5.66 and applied. Booted
> > > without oops. Me happy. :)
> >
> > Nice, thanks for testing, I really appreciate it.
>
> No worries. Glad to help. :)
>
> One point of note though. Should this be happening?
I know I see this on the 2.5, and the original 2.4 i2c code, I think
it's just part of the probing process. The people on the sensor's list
can tell you more, but if the adm1021 driver seems to work, I wouldn't
worry about it :)
thanks,
greg k-h
Thank you very much for merging some of PC-9800 support patches.
I tested quickly using some types of PC-98, desktop, note book and
tower(SMP server). They work fine after applied rest of patch set.
Thanks again,
Osamu Tomita
P.S. I put the rest of patch this URL.
http://downloads.sourceforge.jp/linux98/2842/linux98-2.5.66.patch.tar.bz2
> drivers/char/genrtc.c:100: warning: static declaration for
> `gen_rtc_interrupt' follows non-static
> drivers/char/genrtc.c: In function `gen_rtc_timer':
[...]
It was broken somewhere between -bk1 and -bk4.
Eike
On Tuesday 25th March 2003 at 09:05 uur Rolf Eike Beer wrote:
> It was broken somewhere between -bk1 and -bk4.
Geert Uytterhoeven posted this patch for it, two days ago, it is already
present in 2.5.65-ac4.
--- linux-2.5.x/include/asm-generic/rtc.h.orig Mon Feb 10 21:59:25 2003
+++ linux-2.5.x/include/asm-generic/rtc.h Sun Mar 23 11:47:24 2003
@@ -22,9 +22,8 @@
#define RTC_AIE 0x20 /* alarm interrupt enable */
#define RTC_UIE 0x10 /* update-finished interrupt enable */
-extern void gen_rtc_interrupt(unsigned long);
-
/* some dummy definitions */
+#define RTC_BATT_BAD 0x100 /* battery bad */
#define RTC_SQWE 0x08 /* enable square-wave output */
#define RTC_DM_BINARY 0x04 /* all time/date values are BCD if clear */
#define RTC_24H 0x02 /* 24 hour mode - else hours bit 7 means pm */
@@ -43,7 +42,7 @@
return uip;
}
-static inline void get_rtc_time(struct rtc_time *time)
+static inline unsigned int get_rtc_time(struct rtc_time *time)
{
unsigned long uip_watchdog = jiffies;
unsigned char ctrl;
@@ -108,6 +107,8 @@
time->tm_year += 100;
time->tm_mon--;
+
+ return RTC_24H;
}
/* Set the current date and time in the real time clock. */
--- linux-2.5.x/include/asm-parisc/rtc.h.orig Wed Aug 28 08:33:46 2002
+++ linux-2.5.x/include/asm-parisc/rtc.h Sun Mar 23 11:52:15 2003
@@ -24,7 +24,7 @@
#define RTC_AIE 0x20 /* alarm interrupt enable */
#define RTC_UIE 0x10 /* update-finished interrupt enable */
-extern void gen_rtc_interrupt(unsigned long);
+#define RTC_BATT_BAD 0x100 /* battery bad */
/* some dummy definitions */
#define RTC_SQWE 0x08 /* enable square-wave output */
@@ -44,14 +44,14 @@
{ 0, 31, 60, 91, 121, 152, 182, 213, 244, 274, 305, 335, 366 }
};
-static int get_rtc_time(struct rtc_time *wtime)
+static inline unsigned int get_rtc_time(struct rtc_time *wtime)
{
struct pdc_tod tod_data;
long int days, rem, y;
const unsigned short int *ip;
if(pdc_tod_read(&tod_data) < 0)
- return -1;
+ return RTC_24H | RTC_BATT_BAD;
// most of the remainder of this function is:
@@ -93,7 +93,7 @@
wtime->tm_mon = y;
wtime->tm_mday = days + 1;
- return 0;
+ return RTC_24H;
}
static int set_rtc_time(struct rtc_time *wtime)
--- linux-2.5.x/include/asm-ppc/rtc.h.orig Tue Feb 18 10:08:44 2003
+++ linux-2.5.x/include/asm-ppc/rtc.h Sun Mar 23 11:47:31 2003
@@ -35,15 +35,14 @@
#define RTC_AIE 0x20 /* alarm interrupt enable */
#define RTC_UIE 0x10 /* update-finished interrupt enable */
-extern void gen_rtc_interrupt(unsigned long);
-
/* some dummy definitions */
+#define RTC_BATT_BAD 0x100 /* battery bad */
#define RTC_SQWE 0x08 /* enable square-wave output */
#define RTC_DM_BINARY 0x04 /* all time/date values are BCD if clear */
#define RTC_24H 0x02 /* 24 hour mode - else hours bit 7 means pm */
#define RTC_DST_EN 0x01 /* auto switch DST - works f. USA only */
-static inline void get_rtc_time(struct rtc_time *time)
+static inline unsigned int get_rtc_time(struct rtc_time *time)
{
if (ppc_md.get_rtc_time) {
unsigned long nowtime;
@@ -55,6 +54,7 @@
time->tm_year -= 1900;
time->tm_mon -= 1; /* Make sure userland has a 0-based month */
}
+ return RTC_24H;
}
/* Set the current date and time in the real time clock. */
Dana utorak 25. mart 2003. 04:19 napisali ste:
>
> Its radeonfb now instead of radeon. Alot of drivers where broken in this
> way. Now every driver follows a standard.
>
Nope, same thing with video=radeonfb...
>
> For console resizing try using stty cols xxx rows xx.
>
Tried. Not working again. Last line of the text is at same position like
when changing mode with fbset, upper lines are now on the right where garbage
is when using fbset.
First scrolling gives an oops, but due to screen corruption I could not write
down message displayed. Nothing in logs due to irregular reboot.
> >
> > My config is attached, gcc version is 2.95.3, modutils 2.4.21.
Forgot to add, using GV-R9000 (GigaByte Maya R9000)
--
Pozdrav,
Tanasković Toplica
On Mon, 24 Mar 2003, Linus Torvalds wrote:
> A lot of changes all over. Most notably probably the fbcon updates
* And this changes aren't good for tdfx driver :(
I've got Voodoo3:
01:00.0 VGA compatible controller: 3Dfx Interactive, Inc. Voodoo 3 (rev 01) (prog-if 00 [VGA])
Subsystem: 3Dfx Interactive, Inc. Voodoo3 AGP
Flags: 66Mhz, fast devsel, IRQ 11
Memory at e4000000 (32-bit, non-prefetchable) [size=32M]
Memory at e8000000 (32-bit, prefetchable) [size=32M]
I/O ports at c000 [size=256]
Expansion ROM at <unassigned> [disabled] [size=64K]
Capabilities: [54] AGP version 1.0
Capabilities: [60] Power Management version 1
Only 8-bit works (more or less) fine. 16-bit I have white letters on...
white background :) Very nice for work :D In 2.5.64 I had white letters on
brown-red backgound.
24-bit has also colours: white letter on very light blue.
My screen is completelly broken with green characters and small screen in
corner when I switch from X to console. It was before sometimes, now it's
always :(
* Next thing (sorry Linus for priv letter...):
[...]
if [ -r System.map ]; then /sbin/depmod -ae -F System.map 2.5.66; fi
WARNING: /lib/modules/2.5.66/kernel/drivers/md/xor.ko needs unknown symbol kernel_fpu_begin
WARNING: /lib/modules/2.5.66/kernel/drivers/scsi/aha152x.ko needs unknown symbol scsi_put_command
WARNING: /lib/modules/2.5.66/kernel/drivers/scsi/aha152x.ko needs unknown symbol scsi_get_command
WARNING: /lib/modules/2.5.66/kernel/drivers/hotplug/acpiphp.ko needs unknown symbol acpi_resource_to_address64
#
The most important thing - I have a lot of oops in my logs. ~ 1 per 2
seconds
Debug: sleeping function called from illegal context at mm/slab.c:1723
Call Trace:
[<c011c414>] __might_sleep+0x54/0x60
[<c013b86b>] kmalloc+0x4b/0xa0
[<c02042b0>] accel_cursor+0xe0/0x2d0
[<c011a919>] try_to_wake_up+0x179/0x220
[<c0204678>] fb_vbl_handler+0x78/0xa0
[<c0124e7c>] update_process_times+0x2c/0x40
[<c0124d4d>] update_wall_time+0xd/0x40
[<c0202e40>] cursor_timer_handler+0x10/0x30
[<c0124fa8>] run_timer_softirq+0xf8/0x130
[<c0202e30>] cursor_timer_handler+0x0/0x30
[<c0121801>] do_softirq+0x51/0xb0
[<c010c700>] do_IRQ+0xf0/0x110
[<c010b064>] common_interrupt+0x18/0x20
[<c01c6a3e>] acpi_processor_idle+0xfe/0x210
[<c01c6940>] acpi_processor_idle+0x0/0x210
[<c0109040>] default_idle+0x0/0x30
[<c01090f2>] cpu_idle+0x32/0x40
[<c0105000>] rest_init+0x0/0x60
[<c0105055>] rest_init+0x55/0x60
My config is on URL: http://piorun.ds.pg.gda.pl/~blues/config-2.5.66.txt
--
---------------------------------
pozdr. Pawe? Go?aszewski
---------------------------------
CPU not found - software emulation...
> * And this changes aren't good for tdfx driver :(
Sorry I haven't got around to the driver.
> The most important thing - I have a lot of oops in my logs. ~ 1 per 2
> seconds
Try this patch and let me know how it works.
diff -urN -X /home/jsimmons/dontdiff linus-2.5/drivers/video/console/fbcon.c fbdev-2.5/drivers/video/console/fbcon.c
--- linus-2.5/drivers/video/console/fbcon.c Sat Mar 22 21:45:23 2003
+++ fbdev-2.5/drivers/video/console/fbcon.c Tue Mar 25 12:03:56 2003
@@ -172,8 +172,9 @@
* Internal routines
*/
static void fbcon_set_display(struct vc_data *vc, int init, int logo);
+static void accel_cursor(struct vc_data *vc, struct fb_info *info,
+ struct fb_cursor *cursor, int yy);
static __inline__ int real_y(struct display *p, int ypos);
-static void fb_vbl_handler(int irq, void *dummy, struct pt_regs *fp);
static __inline__ void updatescrollmode(struct display *p, struct vc_data *vc);
static __inline__ void ywrap_up(struct vc_data *vc, int count);
static __inline__ void ywrap_down(struct vc_data *vc, int count);
@@ -194,6 +195,34 @@
}
#endif
+static void fb_callback(void *private)
+{
+ struct fb_info *info = (struct fb_info *) private;
+ struct display *p = &fb_display[fg_console];
+ struct vc_data *vc = vc_cons[fg_console].d;
+ struct fb_cursor cursor;
+
+ if (!info || !cursor_on)
+ return;
+
+ if (vbl_cursor_cnt && --vbl_cursor_cnt == 0) {
+ cursor.set = 0;
+
+ if (!cursor_drawn)
+ cursor.set = FB_CUR_SETCUR;
+ accel_cursor(vc, info, &cursor, real_y(p, vc->vc_y));
+ cursor_drawn ^= 1;
+ vbl_cursor_cnt = cursor_blink_rate;
+ }
+}
+
+static void fb_vbl_handler(int irq, void *dev_id, struct pt_regs *fp)
+{
+ struct fb_info *info = dev_id;
+
+ schedule_work(&info->queue);
+}
+
static void cursor_timer_handler(unsigned long dev_addr);
static struct timer_list cursor_timer =
@@ -203,7 +232,7 @@
{
struct fb_info *info = (struct fb_info *) dev_addr;
- fb_vbl_handler(0, info, NULL);
+ schedule_work(&info->queue);
cursor_timer.expires = jiffies + HZ / 50;
add_timer(&cursor_timer);
}
@@ -290,14 +319,14 @@
const unsigned short *s)
{
unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
- unsigned int width = (vc->vc_font.width + 7)/8;
+ unsigned int width = (vc->vc_font.width + 7) >> 3;
unsigned int cellsize = vc->vc_font.height * width;
unsigned int maxcnt = info->pixmap.size/cellsize;
unsigned int shift_low = 0, mod = vc->vc_font.width % 8;
unsigned int shift_high = 8, size, pitch, cnt, k;
unsigned int buf_align = info->pixmap.buf_align - 1;
unsigned int scan_align = info->pixmap.scan_align - 1;
- unsigned int idx = vc->vc_font.width/8;
+ unsigned int idx = vc->vc_font.width >> 3;
u8 mask, *src, *dst, *dst0;
while (count) {
@@ -307,7 +336,7 @@
cnt = k = count;
image->width = vc->vc_font.width * cnt;
- pitch = (image->width + 7)/8 + scan_align;
+ pitch = ((image->width + 7) >> 3) + scan_align;
pitch &= ~scan_align;
size = pitch * vc->vc_font.height + buf_align;
size &= ~buf_align;
@@ -338,7 +367,7 @@
const unsigned short *s)
{
unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
- unsigned int width = vc->vc_font.width/8;
+ unsigned int width = vc->vc_font.width >> 3;
unsigned int cellsize = vc->vc_font.height * width;
unsigned int maxcnt = info->pixmap.size/cellsize;
unsigned int scan_align = info->pixmap.scan_align - 1;
@@ -411,7 +440,7 @@
int c, int ypos, int xpos)
{
unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
- unsigned int width = (vc->vc_font.width + 7)/8;
+ unsigned int width = (vc->vc_font.width + 7) >> 3;
unsigned int scan_align = info->pixmap.scan_align - 1;
unsigned int buf_align = info->pixmap.buf_align - 1;
int bgshift = (vc->vc_hi_font_mask) ? 13 : 12;
@@ -559,6 +588,15 @@
vc = (struct vc_data *) kmalloc(sizeof(struct vc_data), GFP_ATOMIC);
+ if (!vc) {
+ if (softback_buf)
+ kfree((void *) softback_buf);
+ return NULL;
+ }
+
+ /* Initialize the work queue */
+ INIT_WORK(&info->queue, fb_callback, info);
+
/* Setup default font */
vc->vc_font.data = font->data;
vc->vc_font.width = font->width;
@@ -956,8 +994,8 @@
accel_putcs(vc, info, s, count, real_y(p, ypos), xpos);
}
-void accel_cursor(struct vc_data *vc, struct fb_info *info, struct fb_cursor *cursor,
- int yy)
+static void accel_cursor(struct vc_data *vc, struct fb_info *info,
+ struct fb_cursor *cursor, int yy)
{
unsigned short charmask = vc->vc_hi_font_mask ? 0x1ff : 0xff;
int bgshift = (vc->vc_hi_font_mask) ? 13 : 12;
@@ -986,7 +1024,15 @@
size = ((width + 7) >> 3) * height;
data = kmalloc(size, GFP_KERNEL);
+
+ if (!data) return;
+
mask = kmalloc(size, GFP_KERNEL);
+
+ if (!mask) {
+ kfree(data);
+ return;
+ }
if (cursor->set & FB_CUR_SETSIZE) {
memset(data, 0xff, size);
@@ -1101,27 +1147,6 @@
}
}
-static void fb_vbl_handler(int irq, void *dev_id, struct pt_regs *fp)
-{
- struct fb_info *info = dev_id;
- struct display *p = &fb_display[fg_console];
- struct vc_data *vc = vc_cons[fg_console].d;
- struct fb_cursor cursor;
-
- if (!cursor_on)
- return;
-
- if (vbl_cursor_cnt && --vbl_cursor_cnt == 0) {
- cursor.set = 0;
-
- if (!cursor_drawn)
- cursor.set = FB_CUR_SETCUR;
- accel_cursor(vc, info, &cursor, real_y(p, vc->vc_y));
- cursor_drawn ^= 1;
- vbl_cursor_cnt = cursor_blink_rate;
- }
-}
-
static int scrollback_phys_max = 0;
static int scrollback_max = 0;
static int scrollback_current = 0;
diff -urN -X /home/jsimmons/dontdiff linus-2.5/drivers/video/softcursor.c fbdev-2.5/drivers/video/softcursor.c
--- linus-2.5/drivers/video/softcursor.c Sat Mar 22 21:45:22 2003
+++ fbdev-2.5/drivers/video/softcursor.c Tue Mar 25 11:41:28 2003
@@ -44,6 +44,7 @@
if (info->cursor.mask)
kfree(info->cursor.mask);
info->cursor.mask = kmalloc(dsize, GFP_KERNEL);
+ if (!info->cursor.mask) return -ENOMEM;
if (cursor->mask)
memcpy(info->cursor.mask, cursor->mask, dsize);
else
diff -urN -X /home/jsimmons/dontdiff linus-2.5/include/linux/fb.h fbdev-2.5/include/linux/fb.h
--- linus-2.5/include/linux/fb.h Sat Mar 22 21:45:25 2003
+++ fbdev-2.5/include/linux/fb.h Tue Mar 25 12:00:20 2003
@@ -2,6 +2,7 @@
#define _LINUX_FB_H
#include <linux/tty.h>
+#include <linux/workqueue.h>
#include <asm/types.h>
#include <asm/io.h>
@@ -406,8 +407,9 @@
struct fb_fix_screeninfo fix; /* Current fix */
struct fb_monspecs monspecs; /* Current Monitor specs */
struct fb_cursor cursor; /* Current cursor */
- struct fb_cmap cmap; /* Current cmap */
+ struct work_struct queue; /* Framebuffer event queue */
struct fb_pixmap pixmap; /* Current pixmap */
+ struct fb_cmap cmap; /* Current cmap */
struct fb_ops *fbops;
char *screen_base; /* Virtual address */
struct vc_data *display_fg; /* Console visible on this display */
On Wed, 26 Mar 2003, Geert Uytterhoeven wrote:
> On Tue, 25 Mar 2003 [email protected] wrote:
> > Good! But I don't see my compilation fixes. Will resend.
> > Below a new one.
> >
> > Andries
> >
> > diff -u --recursive --new-file -X /linux/dontdiff a/fs/partitions/msdos.c b/fs/partitions/msdos.c
> > --- a/fs/partitions/msdos.c Mon Feb 24 23:02:56 2003
> > +++ b/fs/partitions/msdos.c Tue Mar 25 06:22:31 2003
> > @@ -219,7 +219,7 @@
> > * Create devices for BSD partitions listed in a disklabel, under a
> > * dos-like partition. See parse_extended() for more information.
> > */
> > -static void
> > +void
> > parse_bsd(struct parsed_partitions *state, struct block_device *bdev,
> > u32 offset, u32 size, int origin, char *flavour,
> > int max_partitions)
>
> Doesn't it make sense to move parse_bsd() to a separate file? Else we have to
> add a dependency on MSDOS_PARTITION to NEC98_PARTITION.
Should have been: ... if BSD_DISKLABEL is set.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
On Tue, 25 Mar 2003 [email protected] wrote:
> Good! But I don't see my compilation fixes. Will resend.
> Below a new one.
>
> Andries
>
> diff -u --recursive --new-file -X /linux/dontdiff a/fs/partitions/msdos.c b/fs/partitions/msdos.c
> --- a/fs/partitions/msdos.c Mon Feb 24 23:02:56 2003
> +++ b/fs/partitions/msdos.c Tue Mar 25 06:22:31 2003
> @@ -219,7 +219,7 @@
> * Create devices for BSD partitions listed in a disklabel, under a
> * dos-like partition. See parse_extended() for more information.
> */
> -static void
> +void
> parse_bsd(struct parsed_partitions *state, struct block_device *bdev,
> u32 offset, u32 size, int origin, char *flavour,
> int max_partitions)
Doesn't it make sense to move parse_bsd() to a separate file? Else we have to
add a dependency on MSDOS_PARTITION to NEC98_PARTITION.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
On Mon, Mar 24, 2003 at 03:26:47PM -0800, Linus Torvalds wrote:
>...
> Summary of changes from v2.5.65 to v2.5.66
> ============================================
>...
> Dave Jones <[email protected]>:
>...
> o guard mad16 debug macro
>...
This change broke the compilation of sound/oss/mad16.c:
<-- snip -->
...
gcc -Wp,-MD,sound/oss/.mad16.o.d -D__KERNEL__ -Iinclude -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -pipe
-mpreferred-stack-boundary=2 -march=k6 -Iinclude/asm-i386/mach-default -nostdinc
-iwithprefix include -DKBUILD_BASENAME=mad16 -DKBUILD_MODNAME=mad16 -c -o
sound/oss/mad16.o sound/oss/mad16.c
...
sound/oss/mad16.c: In function `probe_mad16':
sound/oss/mad16.c:541: syntax error before "else"
sound/oss/mad16.c:604: syntax error before "else"
make[2]: *** [sound/oss/mad16.o] Error 1
<-- snip -->
The following patch is needed:
--- linux-2.5.66-notfull/sound/oss/mad16.c.old 2003-03-26 11:52:13.000000000 +0100
+++ linux-2.5.66-notfull/sound/oss/mad16.c 2003-03-26 11:55:06.000000000 +0100
@@ -537,7 +537,7 @@
for (i = 0xf8d; i <= 0xf93; i++) {
if (!c924pnp)
- DDB(printk("port %03x = %02x\n", i, mad_read(i)))
+ DDB(printk("port %03x = %02x\n", i, mad_read(i)));
else
DDB(printk("port %03x = %02x\n", i-0x80, mad_read(i)));
}
@@ -600,7 +600,7 @@
for (i = 0xf8d; i <= 0xf93; i++) {
if (!c924pnp)
- DDB(printk("port %03x after init = %02x\n", i, mad_read(i)))
+ DDB(printk("port %03x after init = %02x\n", i, mad_read(i)));
else
DDB(printk("port %03x after init = %02x\n", i-0x80, mad_read(i)));
}
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
On Mon, Mar 24, 2003 at 03:26:47PM -0800, Linus Torvalds wrote:
>...
> Summary of changes from v2.5.65 to v2.5.66
> ============================================
>...
> Jaroslav Kysela <[email protected]>:
> o ALSA update
>...
snd-ice1724 seems to be too much of a copy of snd-ice1712, trying to
compile both into the kernel results in the following error:
<-- snip -->
...
ld -m elf_i386 -r -o sound/pci/ice1712/built-in.o
sound/pci/ice1712/snd-ice1712.o sound/pci/ice1712/snd-ice1724.o
sound/pci/ice1712/snd-ice1724.o(.text+0x540): In function
`snd_ice1712_akm4xxx_init':
: multiple definition of `snd_ice1712_akm4xxx_init'
sound/pci/ice1712/snd-ice1712.o(.text+0x7540): first defined here
sound/pci/ice1712/snd-ice1724.o(.text+0x2c0): In function
`snd_ice1712_akm4xxx_reset':
: multiple definition of `snd_ice1712_akm4xxx_reset'
sound/pci/ice1712/snd-ice1712.o(.text+0x72c0): first defined here
sound/pci/ice1712/snd-ice1724.o(.text+0x8c0): In function
`snd_ice1712_akm4xxx_build_controls':
: multiple definition of `snd_ice1712_akm4xxx_build_controls'
sound/pci/ice1712/snd-ice1712.o(.text+0x78c0): first defined here
sound/pci/ice1712/snd-ice1724.o(.text+0x0): In function
`snd_ice1712_akm4xxx_write':
: multiple definition of `snd_ice1712_akm4xxx_write'
sound/pci/ice1712/snd-ice1712.o(.text+0x7000): first defined here
make[3]: *** [sound/pci/ice1712/built-in.o] Error 1
<-- snip -->
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
From: Geert Uytterhoeven <[email protected]>
> -static void
> +void
Doesn't it make sense to move parse_bsd() to a separate file?
Else we have to add a dependency on MSDOS_PARTITION to NEC98_PARTITION.
Yes, it does. But this PC-9800 stuff is still in flux.
Once it has settled down we can see whether reorganization
improves things. This is just the minimal compilation fix.
Andries
On Mon, Mar 24, 2003 at 03:26:47PM -0800, Linus Torvalds wrote:
>...
> Summary of changes from v2.5.65 to v2.5.66
> ============================================
>...
> Christoph Hellwig <[email protected]>:
>...
> o misc devfs_register cleanups
>...
This patch broke the compilation of drivers/char/ipmi/ipmi_devintf.c:
<-- snip -->
...
gcc -Wp,-MD,drivers/char/ipmi/.ipmi_devintf.o.d -D__KERNEL__ -Iinclude
-Wall -Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing -fno-common
-pipe -mpreferred-stack-boundary=2 -march=k6 -Iinclude/asm-i386/mach-default
-nostdinc -iwithprefix include -DKBUILD_BASENAME=ipmi_devintf
-DKBUILD_MODNAME=ipmi_devintf -c -o drivers/char/ipmi/ipmi_devintf.o
drivers/char/ipmi/ipmi_devintf.c
drivers/char/ipmi/ipmi_devintf.c: In function `ipmi_new_smi':
drivers/char/ipmi/ipmi_devintf.c:452: warning: implicit declaration of
function `snprinf'
...
... -o .tmp_vmlinux1
...
drivers/built-in.o(.text+0x142824): In function `ipmi_new_smi':
: undefined reference to `snprinf'
...
make: *** [.tmp_vmlinux1] Error 1
<-- snip -->
Trivial fix:
--- linux-2.5.66-notfull/drivers/char/ipmi/ipmi_devintf.c.old 2003-03-26 13:34:17.000000000 +0100
+++ linux-2.5.66-notfull/drivers/char/ipmi/ipmi_devintf.c 2003-03-26 13:34:34.000000000 +0100
@@ -449,7 +449,7 @@
if (if_num > MAX_DEVICES)
return;
- snprinf(name, sizeof(name), "ipmidev/%d", if_num);
+ snprintf(name, sizeof(name), "ipmidev/%d", if_num);
handles[if_num] = devfs_register(NULL, name, DEVFS_FL_NONE,
ipmi_major, if_num,
cu
Adrian
--
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed
diff -urN linux.orig/Documentation/IPMI.txt linux-main/Documentation/IPMI.txt
--- linux.orig/Documentation/IPMI.txt Tue Jan 14 11:16:08 2003
+++ linux-main/Documentation/IPMI.txt Wed Mar 26 08:05:10 2003
@@ -5,6 +5,18 @@
<[email protected]>
<[email protected]>
+The Intelligent Platform Management Interface, or IPMI, is a
+standard for controlling intelligent devices that monitor a system.
+It provides for dynamic discovery of sensors in the system and the
+ability to monitor the sensors and be informed when the sensor's
+values change or go outside certain boundaries. It also has a
+standardized database for field-replacable units (FRUs) and a watchdog
+timer.
+
+To use this, you need an interface to an IPMI controller in your
+system (called a Baseboard Management Controller, or BMC) and
+management software that can use the IPMI system.
+
This document describes how to use the IPMI driver for Linux. If you
are not familiar with IPMI itself, see the web site at
http://www.intel.com/design/servers/ipmi/index.htm. IPMI is a big
diff -urN linux.orig/drivers/char/ipmi/Kconfig linux-main/drivers/char/ipmi/Kconfig
--- linux.orig/drivers/char/ipmi/Kconfig Tue Jan 14 11:16:10 2003
+++ linux-main/drivers/char/ipmi/Kconfig Wed Mar 26 08:05:10 2003
@@ -7,8 +7,14 @@
tristate 'IPMI top-level message handler'
help
This enables the central IPMI message handler, required for IPMI
- to work. Note that you must have this enabled to do any other IPMI
- things. See IPMI.txt for more details.
+ to work.
+
+ IPMI is a standard for managing sensors (temperature,
+ voltage, etc.) in a system.
+
+ See Documentation/IPMI.txt for more details on the driver.
+
+ If unsure, say N.
config IPMI_PANIC_EVENT
bool 'Generate a panic event to all BMCs on a panic'
diff -urN linux.orig/drivers/char/ipmi/ipmi_devintf.c linux-main/drivers/char/ipmi/ipmi_devintf.c
--- linux.orig/drivers/char/ipmi/ipmi_devintf.c Wed Mar 26 08:00:57 2003
+++ linux-main/drivers/char/ipmi/ipmi_devintf.c Wed Mar 26 08:13:31 2003
@@ -449,7 +449,7 @@
if (if_num > MAX_DEVICES)
return;
- snprinf(name, sizeof(name), "ipmidev/%d", if_num);
+ snprintf(name, sizeof(name), "ipmidev/%d", if_num);
handles[if_num] = devfs_register(NULL, name, DEVFS_FL_NONE,
ipmi_major, if_num,
diff -urN linux.orig/drivers/char/ipmi/ipmi_kcs_intf.c linux-main/drivers/char/ipmi/ipmi_kcs_intf.c
--- linux.orig/drivers/char/ipmi/ipmi_kcs_intf.c Fri Feb 28 16:03:18 2003
+++ linux-main/drivers/char/ipmi/ipmi_kcs_intf.c Wed Mar 26 08:14:11 2003
@@ -826,7 +826,7 @@
if (kcs_port && kcs_physaddr)
return -EINVAL;
- new_kcs = kmalloc(kcs_size(), GFP_KERNEL);
+ new_kcs = kmalloc(sizeof(*new_kcs), GFP_KERNEL);
if (!new_kcs) {
printk(KERN_ERR "ipmi_kcs: out of memory\n");
return -ENOMEM;
> > For console resizing try using stty cols xxx rows xx.
> >
> Tried. Not working again. Last line of the text is at same position like
> when changing mode with fbset, upper lines are now on the right where garbage
> is when using fbset.
> First scrolling gives an oops, but due to screen corruption I could not write
> down message displayed. Nothing in logs due to irregular reboot.
I seen this bug. I fixed it in BK.
I've got a similar bug on my sis framebuffer, using 2.5.66 and the
latest patch you sent to the kernel mailing list Wednesday morning with
the subject "Framebuffer fixes." That was about half an hour before
the message where you said you had fixed this in bk, so maybe the patch
didn't have the fix?
Anyway the bug I see is: If I use fbset to change between 1024x768,
800x600, and 640x480 the console doesn't seem to be aware of the
change. However, it doesn't seem to cause corruption or oops'es, at
least for me so far.
However, I am pleased to say that I am able to use the sis framebuffer
driver now. Last time I tried, around 2.5.64 I think, I got serious
screen corruption switching between X and the framebuffer console.
So things are getting better... thanks for all your work!
Torrey Hoffman
On Wed, 2003-03-26 at 12:24, James Simmons wrote:
> > > For console resizing try using stty cols xxx rows xx.
> > >
> > Tried. Not working again. Last line of the text is at same position like
> > when changing mode with fbset, upper lines are now on the right where garbage
> > is when using fbset.
> > First scrolling gives an oops, but due to screen corruption I could not write
> > down message displayed. Nothing in logs due to irregular reboot.
>
> I seen this bug. I fixed it in BK.
--
Torrey Hoffman <[email protected]>
> Anyway the bug I see is: If I use fbset to change between 1024x768,
> 800x600, and 640x480 the console doesn't seem to be aware of the
> change. However, it doesn't seem to cause corruption or oops'es, at
> least for me so far.
Because the framebuffer can operate independent of the console now.try
using stty for now. Fbset will have to be updated to the latest changes.
> However, I am pleased to say that I am able to use the sis framebuffer
> driver now. Last time I tried, around 2.5.64 I think, I got serious
> screen corruption switching between X and the framebuffer console.
>
> So things are getting better... thanks for all your work!
:-)
While we're on the framebuffer bug train, James, do you know of
this bug with radeonfb:
My 2.5 kernel boots. Some initial boot text with ACPI and such is
scrolled on the screen, this is before radeonfb has taken over and
switched the screen size. But this is usually instant.
Right after the switch there is a lot of random characters in
varying colors at the top of the screen below the penguin. The
first legible boot message I see is this:
"Console: switching to colour framebuffer device 128x48"
(not verbatim)
The junk quickly scrolls off into the sunset and has no adverse
effects on the following boot messages.
It does not help to tell lilo to use 1024x768x16 by default.
(vga=791)
This is a minor bug, but the same thing works fine in the 2.4
radeonfb.
Regards
Josh
--
New PGP public key: 0x27AFC3EE
Dana sreda 26. mart 2003. 21:24 napisali ste:
> > > For console resizing try using stty cols xxx rows xx.
> >
> > Tried. Not working again. Last line of the text is at same position
> > like when changing mode with fbset, upper lines are now on the right
> > where garbage is when using fbset.
> > First scrolling gives an oops, but due to screen corruption I could not
> > write down message displayed. Nothing in logs due to irregular reboot.
>
> I seen this bug. I fixed it in BK.
Nope, not fixed in BK3 either. Same thing is happening with, and without BK
patch.
--
Pozdrav,
Tanasković Toplica